Upending Stock Market Structure Using Secure Computation
summary
Proposes replacing the human specialist with secure multi-party computation to run a historical-style clearing-price auction so bids stay hidden; uses Gaussian equilibrium models (building on Grossman–Stiglitz and a 2004 Cornell model) to show prices reveal an informed trader’s signal only partially, preserving research incentives. Designs an MPC protocol among brokers/trusted parties that reveals the total number of orders but not individual bids, highlights leakage vs. efficiency tradeoffs, and suggests piloting on S&P futures.
AI-written summary · about
chapters (7)
- 0:00Upending Stock Market Structure Using MPC
- 0:33Talk Outline What exactly is the purpose of the stock market?
- 4:19Stock Market Purpose Price Discovery Important for investment decisions Effectively centralizes distributed information Liquidity - it is a market! NYSE vs NASDAQ (Double Auctions)
- 8:00Walrasian Equilibrium (1877) N resources and Magents Each agent i has an endowment of resource : e Each agent has a utility function (of their wealth)
- 11:07Walrasian Equilibrium Wairasian Equilibrium of Economy consists of a price vector p and a redistribution such that Each agent maximizes their utility
- 14:38Grossman-Stiglitz Critique (1980) Impossibility of Informationally efficient markets Efficient Markets Hypothesis - Prices reflect all the information Even "Informed" Traders "Private" Information?
- 16:28Easley-O'Hara Model (2004)
related talks
26mSecure Computation in Action: Real-World Deployments in Finance
17mSecure Vickrey Auctions for Online Advertising
01h 31mIntroduction to Secure Multi-Party Computation
34mPrivacy Preserving Aggregation of Ad Conversions using MPC
49mCompiling to preserve our privacy
08mThe Trustless Trade Supply Chain
37mPETs of the world unite! Conquer the middle ground in the performance vs...
57mExtortion and Cooperation in the Zero-day Market